Heat Recovery in Combination with Different Heat Pump Solutions for Energy Supply

(1)

Heat Recovery in Combination with

Different Heat Pump Solutions for Energy Supply

Atle Solberg

Master of Energy Use and Energy Planning Supervisor: Hans Martin Mathisen, EPT Co-supervisor: Maria Justo Alonso, SINTEF

Department of Energy and Process Engineering Submission date: June 2015

Norwegian University of Science and Technology

(2)

(3)

(4)

(5)

Abstract

The main purpose of this Master’s thesis has been to investigate the performance of different methods of heat recovery from ventilation air. Comparisons have been made with regard to delivered energy for heating of domestic hot water (DHW), space heating and ventilation heating. A single-unit dwelling was used as a basis for the simulations.

The house, built in accordance with the Norwegian passive house standard, had a gross internal area of 172.6 m². Seven different combinations of a heat wheel, an exhaust air CO2 heat pump and an outdoor air CO2 heat pump was assessed. Direct–acting electricity was used to cover the remaining heat demand in all seven heat recovery models.

Three different occupant behaviour models and three different climates (Oslo, Stavanger and Kautokeino) were investigated in order to increase the area of application for the results. The building was simulated in IDA Indoor Climate and Energy, while the heat pump performance was post-processed in Matlab. Several aspects concerning heat recovery, dimensioning, ventilation and heating schedules, heat distribution, etc., have been assessed.

Simulations performed for a family of five with a high consumption of domestic hot water in Oslo showed that the required amount of delivered energy for heating was equal to 115 kWh/m²yr, without any heat recovery. The most used heat recovery method in Norway today, a heat wheel, reduces the energy demand by 34%, down to 76 kWh/m²yr.

The most efficient heat recovery model that was found in this report showed a reduction of delivered energy for heating of 68%, down to 36 kWh/m²yr. This heat recovery model consists of a heat wheel and a heat pump utilizing both ventilation exhaust air and outdoor air. The recommended solution for occupants with a high DHW consumption, when economy is taken into account, is based on a heat wheel and an exhaust air CO2heat pump. The amount of delivered energy is reduced by 66%, down to 40 kWh/m²yr.

Any further work within this field should look into the choice of refrigerant, improvements on the heat pump model and the impact of different building sizes. A more in-depth investment analysis of the different heat recovery models, including the heat distribution system in the building, should also be assessed.

III

(6)

(7)

Sammendrag

Formålet med denne masteroppgaven har vært å undersøke ytelsen til ulike metoder for gjenvinning av ventilasjonsvarme. De ulike løsningene har blitt sammenliknet ut ifra mengden levert energi til bygningen for varmt tappevann, romoppvarming og ventilasjonsvarme. En enebolig på 172.6 m², bygd i henhold til den norske passivhusstandarden, ble brukt som utgangspunkt for simuleringene. Syv ulike kombinasjoner av en roterende varmeveksler og CO2 luft–vann varmepumpe ble sett nærmere på. Varmepumpen kunne bruk avtrekksluft og/eller uteluft som varmekilde. Direktevirkende elektrisitet er brukt til å dekke den energien som varmepumpen ikke klarer å dekke. Tre ulike bruker- modeller og tre ulike klimaer (Oslo, Stavanger og Kautokeino) ble undersøkt for å øke anvendelsesområdet til resultatene. Bygningen ble simulert i IDA Indoor Climate and Energy, mens varmepumpeytelsen ble postprossesert i Matlab. Flere ulike aspekter som angår varmegjenvinning, dimensjonering, varmedistribusjon, og tidsstyrte ventilasjons- og temperatursettpunkt ble undersøkt.

Simuleringer ble blant annet utført for en familie på fem med et høyt varmtvannsforbruk, boende i Oslo. Den nødvendige leverte energien for varme til bygningen tilsvarte 115 kWh/m²år, dersom bygningen ikke anvendte varmegjenvinning. Den mest brukte løsningen i Norge i dag, en roterende varmeveksler, reduserte energibehovet med 34%, ned til 76 kWh/m²år. Den mest effektive metoden som ble funnet i denne rap- porten kunne vise til en reduksjon på 68%, ned til 36 kWh/m²år. Denne løsningen består av en roterende varmeveksler og en varmepumpe som kan utnytte både avtrekksluft og uteluft. Den anbefalte løsningen for beboere med et høyt varmtvannsforbruk, når økonomi er tatt med i betraktningen, er basert på en roterende varmeveksler og et avtrekks- varmepumpe. Behovet for levert energi for å dekke varmebehovet er redusert med 66%, ned til 40 kWh/m²år.

Videre arbeid innen dette temaet bør se på valg av arbeidsmedie, forbedringer på varmepumpemodellen, og ulike bygningsstørrelsers påvirkning på valg av varmegjenvinningsmetode. Dessuten bør det gjennomføres en mer dyptgående investeringsanalyse av de ulike varme- gjenvinningsmetodene. En slik analyse bør også se på kostnader for varmedistribusjonssystemet i bygningen.

V

(8)

(9)

Preface

This Master’s Thesis is the final work of the master’s degree in Energy Use and Energy Planning at the Norwegian University of Science and Technology in Trondheim, Norway. The thesis is a continuation of the project thesis written during the autumn 2014.

The scope of the Master’s Thesis is to assess different combinations of ventilation heat exchangers, and heat pumps for ventilation heat recovery and energy supply to the building. The assignment was proposed by Maria Justo Alonso, research scientist at SINTEF Buildings and Infras- tructure in Trondheim. The assignment is connected to the IEA Heat Pump centre Annex 40 – Heat pump concepts for Nearly Zero-Energy Buildings. The thesis comprises 30 ECTS credits.

I would like to thank my supervisors, Maria Justo Alonso and professor Hans Martin Mathisen, for valuable support, feedback and advises during my work on this report. I would also like to thank the employees at the Department of Energy and Process Engineering, and the employees at SINTEF Energy Research, who have been available for discussions and counselling. Especially Associate Professor Natasa Nord, together with Senior Engineer Inge Håvard Rekstad, have provided valuable information regarding IDA ICE and CO2 compressors respectively.

At last, I would like to thank my girlfriend and wife-to-be, Linn Therese, for her love and support during the work on my project and Master’s Thesis.

Atle Solberg Trondheim, 11th of June 2015

VII

(10)

(11)

B.1 Main script (main.m) . . . 145 B.2 Script containing parameters (parameters.m) . . . 158 B.3 Script containing constants (constants.m) . . . 160 B.4 Function with thermodynamics (thermodynamics.m) . . 161 B.5 Script with preallocations (preallocations.m) . . . 163 B.6 Script clearing parameters (clear_var.m) . . . 165 B.7 Function for isentropic compressor efficiency (eta_is.m) 167 B.8 Function for volumetric compressor efficiency (eta_vol.m)168

(15)

List of Figures

Chapter 2: Theoretical background of heat recovery 3 2.1 Illustration of a rotating wheel heat exchanger . . . 5 2.2 Illustration of the most basic components in a heat pump 13 2.3 T-h diagram showing a general transcritical heat pump

cycle . . . 14 2.4 Comparison of average temperature during heat rejection 15 2.5 Transcritical pressure lines for CO2 . . . 16 2.6 System configuration sketch of air–air/water heat pump 18 2.7 Temp. diff. between discharge air and outdoor air for

different efficiencies . . . 19 2.8 Energy flows in an exhaust air heat pump, according to

NS 3031 (2014) . . . 21

Chapter 3: Literature 23

3.1 System solution of NIBE F470 (Solsem, 2015a). . . 33 3.2 System solution of Genvex 185 S/LS (Solsem, 2015a) . . 35 3.3 System solution of Nilan Compact P Nordic . . . 37 3.4 System solution of Exvent Greenair HP . . . 38

Chapter 4: Basis for simulations 41

4.1 Picture of ”Karita” . . . 42

Chapter 5: Heat recovery simulations in IDA ICE and

Matlab 51

5.1 Architecture of the simulation model . . . 54 5.2 Floor plan of the building ”Karita” . . . 56 5.3 Shading objects from nearby buildings illustrated . . . . 57 5.4 DHW distribution schedule used in the IDA ICE model 60 5.5 Simplified sketch of system solution simulated inMatlab 67 5.6 Isentropic and volumetric compressor efficiencies as a

function of pressure ratio . . . 71 XIII

(16)

XIV List of Figures

5.7 Example of how the half-interval search algorithm works 76

Chapter 6: Results 79

6.1 Heating energy budget for ”NS3031” occupant behaviour model . . . 80 6.2 Duration curve for space heating and ventilation pre-

heating power for Oslo climate . . . 81 6.3 Duration curve for space heating and ventilation heating

power for Stavanger climate . . . 82 6.4 Duration curve for space heating and ventilation heating

power for Kautokeino climate . . . 82 6.5 Duration curve for temperature at the bottom of DHW

tank . . . 84 6.6 Duration curve for DHW boiler power . . . 85 6.7 Optimum compressor volume for ”Large” occupant be-

haviour model (Oslo) . . . 87 6.8 Optimum compressor volume for ”Large” occupant be-

haviour model (Kautokeino) . . . 89 6.9 Delivered energy for heating at different UA values . . . 91 6.10 Delivered energy for heating at different UA values for

different climates . . . 92 6.11 Outdoor air fan volume flow influence on delivered en-

ergy for heating . . . 93 6.12 Influence of DHW storage tank volume on delivered energy 95 6.13 Influence of reduced air flow rates when unoccupied . . 99 6.14 Heat wheel efficiency influence on delivered energy . . . 100 6.15 Sensitivity of isentropic efficiency and compressor heat

loss . . . 101 6.16 Sensitivity of the temperature approach value . . . 103 6.17 Significanse of minimum compressor volume (Oslo, ”Large”)104 6.18 Significanse of minimum compressor volume (Stavanger,

”Small”) . . . 105 6.19 Comparison of solutions with and without outdoor air

as additional heat source (”Small”) . . . 107 6.20 Comparison of solutions with and without outdoor air

as additional heat source (”Large”) . . . 108

(17)

List of Figures XV

6.21 Comparison of heat recovery cases (”Small”) . . . 109

6.22 Comparison of heat recovery cases (”NS3031”) . . . 110

6.23 Comparison of heat recovery cases (”Large”) . . . 111

6.24 Composition of delivered energy for heating . . . 112

6.25 Composition of delivered energy for heating for different climates . . . 113

6.26 The performance ofHWeHP covering DHW only . . . . 115

6.27 COP duration curves for ”Large” occupant behaviour model . . . 116

6.28 Maximum permissible investment (MPI), Oslo climate . 117 6.29 Maximum permissible investment (MPI), Stavanger climate . . . 118

(18)

(19)

List of Tables

Chapter 2: Theoretical background of heat recovery 3 2.1 Comparison of different heat/energy exchangers . . . 11 2.2 Minimum discharge air temperature from heat exchangers 11

Chapter 3: Literature 23

3.1 Test results for three different commercialized products. 29 3.2 Comparison of the different heat recovery products . . . 40

Chapter 4: Basis for simulations 41

4.1 Thermal properties and energy efficiencies of ”Karita” . 42 4.2 Standard figures from NS 3031 (2014) for DHW, lighting

and equipment. . . 44 4.3 Calculation of DHW consumption for ”Small” occupant

behaviour model . . . 45 4.4 Calculation of DHW consumption for ”Large” occupant

behaviour model . . . 46 4.5 Comparison of energy posts for the three occupants be-

haviour models . . . 46 4.6 Overview of the different heat recovery cases in the report 48 4.7 Climate data used in the simulations . . . 49

Chapter 5: Heat recovery simulations in IDA ICE and

Matlab 51

5.1 Deviations from default tank and boiler parameters for DHW system . . . 59 5.2 Deviations from the default tank and boiler parameters

for SH and VH system . . . 61 5.3 Deviations from default air handling unit setup in IDA ICE 64 5.4 Description of the differentMatlabfiles . . . 66 5.5 Optimum CO2 gas cooler pressure according to Stene

(2014) . . . 73 XVII

(20)

XVIII List of Tables

Chapter 6: Results 79

6.1 Comparison of optimum compressor volume flow at full capacity for Oslo and Kautokeino climate . . . 90 6.2 Influence of night time set-back on delivered energy . . . 96 6.3 Influence of reduced air flow rates when unoccupied . . 98 6.4 All heat recovery solutions compared to no heat recovery 106 6.5 Input parameters for economical calculations . . . 117 6.6 MPI of hydronic heat distribution system . . . 119

(21)

Abbreviations

AAHP Air-air heat pump

AHU Air handling unit

ASHRAE American Society of Heating, Refrigerating and Air Conditioning Engineers

AWHP Air-water heat pump

CFC Chlorofluorocarbons (group of refrigerants)

COP Coefficient of performance

CVHD Compact ventilation and heating device

DHW Domestic hot water

EAHP Exhaust air heat pump

GC Gas cooler

GIA Gross internal area (BRA in Norwegian).

The total floor area within the building envelope, excluding external walls, including area occupied by internal walls, as defined in NS 3940 (2012).

GWP Global warming potential

HFC Hydrofluorocarbons (group of refrigerants) HVAC Heating, ventilation and air conditioning

IAQ Indoor air quality

LCA Life–cycle assessment

LCC Life–cycle cost

LMTD Logarithmic mean temperature difference XIX

(22)

XX List of Tables

MPI Maximum permissible investment

NTNU Norges teknisk-vitenskapelige universitet (Norwegian University of Science and Technology)

RCHP Reverse–cycle heat pump. Heat pumps where refrigerant can be reversed using a four-way valve. Can supply heating and cooling.

SFP Specific fan power [kW/(m³/s)]. Power used to move 1 cubic meter of air per second through the ventilation system.

SPF Seasonal performance factor. Heating out- put compared to the total power input over the season, including internal heat exchangers, fans and controls.

SGHE Suction gas heat exchanger, also called internal heat exchanger.

SH Space heating

SINTEF Stiftelsen for industriell og teknisk Forskn- ing (The Foundation for Scientific and In- dustrial Research)

VH Ventilation heating. (Pre)heating of supply air.

ZEB Zero Emission Building

(23)

Glossary

City water Water supplied to the building from the waterworks.

Delivered energy Delivered energy to the building (e.g. electrical energy from the grid).

Discharge air Ventilation air leaving the building to the surrounding environment.

Exhaust air Ventilation air removed from zones in the building.

Expansion device Device used in heat pumps to feed the evaporator with the correct amount of refrigerant. Maintains pressure difference between evaporator and condenser. Usually a valve, but ejectors (commercialized) and expanders (under development) are also possible.

Heating Heating, unless stated otherwise, is heating of DHW, space heating and ventilation heating.

Hot air heating Heating of a building through the supply air. Supply air heated to 30°C to 40°C before entering the zone of occupancy.

Net energy demand Amount of energy needed in within the zones of the building. Efficiency of plant etc. not accounted for.

Outdoor air Air from the surrounding environment entering the building.

Supply air Ventilation air supplied to the zone of occupancy.

XXI

(24)

XXII List of Tables

UA value U value (overall heat transfer coefficient) multiplied with heat exchanger area [W/K].

(25)

Nomenclature

Latin letters

d Diameter [m]

s Specific entropy [kJ/kgK]

v Specific volume [m³/kg]

h Specific enthalpy [kJ/kg]

Q Thermal energy [kWh]

Q˙ Thermal energy per time unit [W or kW]

T Temperature [°C]

V˙ Volume flow [m³/s]

W Compressor energy [kWh]

W˙ Compressor power [W or kW]

Greek letters

η Efficiency [–]

π Pressure ratio in heat pump (high pressure

divided by low pressure)

XXIII

(26)

(27)

Chapter 1

Introduction

In modern, air tight buildings, replacement of the contaminated indoor air with outdoor air is of outmost importance in order to achieve satisfactory indoor air quality. This is normally done using balanced, mechanical ventilation. Replacement of the consumed indoor air with outdoor air leads to a considerable heat loss in cold climates. In developed countries, primary energy used in buildings amounts to about 40% of the total energy consumption (IEA, nd). 50% of the energy consumed in buildings in many developed countries is consumed by the heating, ventilation and air conditioning systems (Calay and Wang, 2013). It is therefore very important to have efficient systems for heat recovery from ventilation air.

The prevalent Norwegian building code states that the heat recovery unit in residential buildings should have a minimum temperature efficiency of 70% during a year, using the energy measures method (Technical Regulations, 2010). In the passive house standard for residential buildings, the minimum temperature efficiency is 80 % during a year (NS 3700, 2013). Even with such high requirements for heat recovery, energy for heating of ventilation air makes out a large part of the total energy budget, together with heating of domestic hot water.

In most new Norwegian buildings, heat recovery is performed using a heat wheel integrated in the air ventilation unit. However, a heat pump could also be used to recover ventilation heat, and at the same time provide heat for space heating and domestic hot water. This report

1

(28)

2 Chapter 1. Introduction

aims to find the best solution for ventilation heat recovery using a heat wheel and/or a heat pump, for single unit passive house dwellings.

Several different combinations are possible. The conclusions are based on simulations carried out on a residential passive house with varying climate and occupant behaviour models.

The results presented in the report are mostly based on delivered energy to the building for heating of domestic hot water, space heating and ventilation heating. The economy of the different heat recovery methods has not been focused on, though some comparisons have been made with regard to the highest permissible investment cost (MPI).

The beginning of the report gives a theoretical introduction to ventilation heat recovery. This includes an overview of the most common heat exchangers, together with information on CO2 heat pumps and system configurations. The definitions regarding heat recovery using heat pumps are also explained. This chapter is partially based on a project report written during the autumn 2014 (Solberg, 2014).

Thereafter, relevant research literature of heat pump heat recovery and hybrid heat recovery is presented and discussed. The amount of relevant research literature within this field is limited. Existing commercialized products for heat pump heat recovery and hybrid heat recovery are also presented and compared.

The next chapters describe the basis for simulations, the simulation platform and architecture of the simulations. This include defining the different heat recovery cases, occupant behaviour, climates, and so on.

The IDA ICE building model is described, together with the developed Matlab heat pump model.

The simulation results are presented in the next chapter, where they are also discussed, together with appurtenant tables and figures.

The result chapter is divided into five sections: The main results from the building simulations are presented. Thereafter, parameters used to design the building and heat pump model are assessed. The sensitivity for some assumed numbers are analysed in the following section. The two last sections look at the general performance of the different heat recovery cases and some economical aspects. A more general discussion is presented in the following chapter. Thereafter comes a conclusion, in addition to suggestions for further work within this field.

(29)

Chapter 2

Theoretical background of heat recovery

§13-1 in Technical Regulations (2010) states that all buildings should have satisfactory indoor air quality (IAQ). Satisfactory IAQ is important for the occupants’ health an well-being, and is achieved by substituting contaminated air with fresh air. This will, in cold climates such as in Norway, result in an inevitable ventilation air heat loss. Thus, building regulations in countries with a cold climate often demand some sort of heat recovery from ventilation air with a minimum efficiency.

In Norway, using the ”Energy measure” method, the requirement is 70% for residential buildings (Technical Regulations, 2010). The yearly minimum temperature efficiency in the residential passive house standard NS 3700 (2013) is 80%.

Ventilation air heat recovery can be achieved using several different methods, though plate heat exchangers and heat wheels are the most common methods in Norway. Heat pumps are extensively used in Swe- den for heat recovery from exhaust ventilation systems. However, heat pumps can also be used in balanced ventilation systems, preferably together with a passive heat exchanger. This chapter will look into the theory of the most common heat recovery solutions, with most atten- tion on heat wheels and air–water heat pumps, and also a combination of the two.

3

(30)

4 Chapter 2. Theoretical background of heat recovery

2.1 Recovery using heat exchangers

Heat recovery using heat exchangers implies using heat exchangers to directly transfer heat from warm air (exhaust air) to cold air (supply air). This is done with no or a very small consumption of energy.

There are several different types of heat exchangers available. They are therefore often divided into two groups distinguished by how they work: Regenerative heat exchangers are cyclic, while recuperative heat exchangers are static. Both types are discussed in the upcoming chapters.

2.1.1 Regenerative heat exchangers

The general working principle of regenerative heat exchangers are al- ternating exposure of heat accumulating surfaces to warm exhaust air and cold supply air (outdoor air). Thus, the supply and exhaust air ducts must be adjacent. An important property concerning regenerative heat exchangers are their ability to transfer latent heat, in addition to sensible heat. The main advantages are the high efficiency, and no or only limited problems related to frost formation on heat exchanger surfaces. The main disadvantage is an inevitable small transfer of air from exhaust air to supply air, which will contain contaminants. The most common types of regenerative heat exchangers are the rotating wheel and the fixed matrix heat exchanger.

Fixed matrix

The fixed matrix heat exchanger contains two different heat accumulating chambers. Supply air flows through one chamber, while exhaust air flows through the second chamber. The champers have a large surface in order to improve the heat transfer between the chamber surface and the air flowing through. After a given time, dampers see to that the air flow in the two chambers are switched. Cold supply air is now flowing through the chamber that was heated by the exhaust air, and vice versa. The dampers usually oscillate every minute, controlled by a timer. Transfer of contaminants and odours are a disadvantage with this heat exchanger type, though to a smaller extent than for the rotat-

(31)

2.1. Recovery using heat exchangers 5

ing wheel. The heat recovery efficiency can be controlled by changing the settings of the timer controlling the dampers (Novakovic et al., 2007).

Rotating wheel

Rotating wheels go by many names (e.g. thermal wheel, energy wheel, enthalpy wheel, heat wheel, desiccant wheel, dehumidification wheel), where the main working principle is the same. However, differences in material and area of application separates the different subtypes of the rotating wheel. This type of heat exchanger is widely used in newer residential and non-residential buildings in Norway due to its efficiency and its good properties regarding frost formation.

A wheel, covering the entire cross-section of the supply air duct and the exhaust air duct, is rotated by a small electric motor. The wheel consists of many small corrugated ducts allowing the air to pass through. Heat is transferred from the air to the surface of the ducts.

As can be seen on figure 2.1, the ducts will alternate between being exposed to hot air (accumulating heat) and cold air (rejecting heat).

The efficiency of the rotating wheel may be adjusted by controlling the rotational speed of the wheel.

Figure 2.1: Illustration of a rotating wheel heat exchanger

(32)

The ducts in the rotating wheel can be made of a hygroscopic material, and the heat exchanger is then called ”energy wheel” or ”enthalpy wheel”. In that case, the energy wheel will be able to transfer sensible and latent heat. Rotating wheel ducts made of a non-hygroscopic material can only transfer moisture if moist air condenses on the cold duct surface and is subsequently evaporated on the warm side. This means that humidity can be recovered from the exhaust air, even without a hygroscopic material. This type of heat exchanger is usually called

”heat wheel”. Heat wheels are the most common solution in modern ventilation units, according to Novakovic et al. (2007). Energy wheels are less used due to higher transfer of odour from exhaust air to supply air, compared to heat wheels, according to Professor Hans Martin Mathisen at NTNU. The odour is transferred with the moisture. How- ever, thermal wheels can be used beneficially in extremely cold climates due to reduced frost formation on the wheel surfaces. The physical di- mensions of rotating wheels are rather small, which is a large advantage for use in small compact ventilation units.

The efficiency, depending on air flows compared to size, is usually between 75% and 85%. Petersen et al. (2009) reports that the temperature efficiencies claimed by manufacturers are about the same as in reality, based on measurements on rotating wheels in Norwegian schools.

Heat/energy wheels have only limited or no problems with frost formation. Energy wheels are less troubled than heat wheels. The frosting process is slow, and the indoor moisture production varies during the day (Alonso et al., 2015). Thus, some frosting on surfaces may occur during hours with high moisture production, but this frost will normally defrost at hours with less humidity production. Problems regarding fouling in the heat exchanger are rather small seeing that the flow changes direction in the small wheel ducts several times a minute.

The major drawback is the air leak from exhaust air to supply air.

Rotating heat exchangers usually have a purge sector (see figure 2.1) used to minimize the problem. Exhaust air trapped in small ducts when entering the fresh air sector is flushed back to the exhaust air duct, due to the purge sector. An inspection of rotating wheels in Norwegian schools revealed a leakage of 0.2% to 0.4% (Petersen et al., 2009).

(33)

2.1.2 Recuperative heat exchangers

Recuperative heat exchangers are static, and transfer heat directly from warm air to cold air. This is usually done by air flows separated by a plate or tube with high thermal conductivity. Heat can also be transferred using a third fluid for transport of energy, which is favourable in cases where the exhaust air duct and the supply air duct are not adjacent.

No air can be carried over from exhaust air to supply air, as long as the heat exchanger is free of defects. This is the main advantage for this type of heat exchangers, seeing that they can be used in highly contaminated zones. However, frost formation in cold climates is a large disadvantage for this heat exchanger type. This problem can be solved by preheating the outdoor air before passing the heat exchanger.

The cost is reduced efficiency and thus increased energy consumption.

The most commonly used recuperative heat exchanger is the plate heat exchanger, though the run-around type is also used.

Plate/tube

Plate heat exchangers consist of several layers of corrugated plates, where the corrugated plates increase the surface to volume flow ratio.

In the gaps between the plates, supply air and exhaust air are flowing alternately. The flow direction of the two air flows are usually in opposite directions (counter-flow) or at a right angle (cross-flow). The pressure loss in these heat exchangers is low (Alonso et al., 2015). This type of heat exchanger is troubled with frost formation. Warm and humid exhaust will condense and freeze on surfaces with a temperature below 0 °C. Frost will eventually block the air flow through the heat exchanger on the exhaust air side and hamper heat transfer. Frost formation problems may be solved by preheating the outdoor air or bypassing some of the outdoor air, at the expense of recovery efficiency.

The efficiency of the heat exchanger can be controlled by bypassing a share of the supply air.

Tube heat exchangers are much the same as plate heat exchangers.

Instead of being separated by plates, the two air flows are separated by tubing. Supply air flows through tubes, while the exhaust air flows

(34)

on the outside of the tubes. Thus, it has a lower risk of freezing (No- vakovic et al., 2007). Efficiency and frost protection mechanisms can be controlled similarly to the plate heat exchanger.

Run-around

Run-around heat exchangers have one large advantage: The supply air duct can be placed remotely from the exhaust air duct (to some extent).

One heat exchanging coil is placed in the supply air duct and another in the exhaust air duct, connected by pipes. A heat transfer fluid (usually a glycol—water mixture) is flowing in the closed circuit. Heat is given off from the exhaust air to the heat transfer liquid flowing in the pipes.

Heat is given off from the heat transfer liquid to the supply air in the supply air duct. The heat transfer liquid can be self-circulating due to density differences or driven by a circulation pump.

Efficiency can be controlled by installing a three-way valve in the liquid circuit. This valve can also be used for avoiding frost formation.

This heat exchanger type has a rather low efficiency. However, a heat pump using the heat transfer liquid as heat source can easily be used to improve the total heat recovery efficiency.

Heat pipe

Heat pipe heat exchangers share the working principle of the run- around heat exchanger. However, they use heat pipes to transfer heat between the ducts. A pipe is placed between the exhaust air and the supply air. A fluid on the inside of the pipe, usually some kind of refrigerant, is evaporated by the warm exhaust air. Due to the reduced density, it moves towards the top of the pipe. The top of the pipe is placed in the supply air duct. Heat is transferred to the supply air from the refrigerant, which condenses and flows down to the bottom of the pipe where it is once more evaporated. Temperature efficiency and frost protection mechanisms can be controlled by bypassing some of the supply air.

(35)

2.1.3 Membrane energy exchangers (MEE)

Membrane energy exchangers make use of a semi-permeable membrane material which allows for moisture in the air to pass, but not air itself.

Thus, both sensible and latent energy can be recovered. This type of energy exchangers are rather new, and much effort has recently been put into research within this field from research institutions in mainly Canada, China and Italy (Abdel-Salam et al., 2014). Differences in temperature and moisture content is the motive power of heat and mass transfer. Several types of membrane energy exchangers have been developed. Two of the most promising solutions are Flat plate energy exchanger andRun-around membrane energy exchanger.

Flat plate

The flat plate energy exchanger has the same size and shape as the plate heat exchanger. However, the materials separating the air flows are made of a semi-permeable membrane material. Heat and moisture are transferred from the exhaust air to the supply air.

Run-around membrane energy exchanger (RAMEE)

The build-up of this energy exchanger is the same of the run-around heat exchanger. Instead of air–water heat exchangers placed in the supply and exhaust air duct, the RAMEE energy exchanger has liquid-to- air membrane energy exchangers (LAMEE). A LAMEE is a membrane plate exchanger with air and a desiccant solution flowing alternately (Abdel-Salam et al., 2014). Heat and moisture are transferred from the exhaust air to the desiccant solution in the LAMEE. The desiccant solution circulates to the supply air duct, where heat and moisture are given off to the supply air, before circulating back to the exhaust air LAMEE.

2.1.4 Comparison of heat exchangers

The different kinds of heat/energy exchangers have different advantages and disadvantages. The heat recovery methods made mention of in this chapter are compared in table 2.1.

(36)

Three out of eight heat exchangers do not demand adjacent ducts.

These solutions are convenient when retrofitting balanced ventilation in older buildings. However, adjacent ducts are not a challenge when building a new residential house.

All of the regenerative heat exchangers have high sensible efficiencies, and experience no or only limited problems related to frost formation. The energy wheel has the highest latent heat recovery efficiency.

However, all three heat exchangers are, to some extent, troubled with transfer of odour from exhaust air to supply air.

The traditional recuperative heat exchangers, where membrane is note made use of, experience large troubles with regard to frost formation in cold climates. Even though the maximum efficiency for a plate heat exchanger can be as high as 90%, the yearly efficiency will drop when measures are taken to avoid frost formation during cold periods.

Thus, these heat exchangers are not suited for use in cold climates.

Table 2.2 from NS 3031 (2014) lists the minimum discharge air temperature for different heat exchangers. Recuperative heat exchangers (not membrane based) in residential buildings have a frost protection temperature of +5 °C to +9 °C, which is well above the freezing point.

Some parts of the heat exchanger, called ”cold corners”, may have a temperature which is somewhat lower than the average temperature.

Thus, it is not enough to have a minimum leaving air temperature just barely above the freezing point.

The membrane based heat exchangers show promising figures, but are not yet commercialised. The heat wheel is the most used heat exchanger for residential buildings, and non-residential buildings where a small leakage from exhaust air to supply air can be accepted. Its properties with regard to frost formation and efficiency are also satisfactory.

Thus, this is the most interesting case to look into in the heat recovery simulations.

(37)

Table 2.1: Comparison of different heat/energy exchangers, based on information from Alonso et al. (2015) unless stated otherwise.

Type of

heat exchanger Adjacent ducts Sensible efficiency Latent efficiency Frost formation Transfer ofodour Regenerative:

Fixed matrix¹ 3 70−80 – Minor 3

Heat wheel 3 50−80 – Minor 3

Energy wheel 3 80−85 80−85 7 3

Recuperative:

Plate/tube 3 60−90 – 3 7

Run-around 7 65−70 – 3 7

Heat pipe¹ 7 50−60 – 3 7

Membr. plate 3 80−85 46−76 7 7

RAMEE 7 60−80 50−65 7 7

1Novakovic et al. (2007).

Table 2.2: Minimum allowed discharge air temperature from heat exchanger, according to NS 3031 (2014)

Heat exchanger type Min. temp.

°C Rotating wheel and fixed matrix, all buildings −10 °C Recuperative (plate, etc.), non-residential +0 °C Recuperative (plate, etc.), residential (optimum) +5 °C Recuperative (plate, etc.), residential (conservative) +9 °C

(38)

2.2 Heat recovery using exhaust air heat pumps

Heat pumps are the most common alternative to heat exchangers for ventilation heat recovery. In addition to recovery of ventilation heat, the heat pump can also supply additional energy to the building. The heat pump may use for example outdoor air and/or energy from the ground as additional heat source, but this is not considered as heat recovery from ventilation air (see chapter 2.4, p. 20).

Using a heat pump for heat recovery has both advantages and disadvantages compared to the heat exchangers mentioned in chapter 2.1.

The supply air and exhaust air duct do not have to be adjacent, which can be an important aspect when retrofitting heat recovery in existing buildings.

Depending on the system solution, the heat pump is able to supply energy for both DHW, space heating and preheating of the supply ventilation air. The heat pump evaporator is usually able to remove more energy from the exhaust air, compared to a heat exchanger. Small, residential passive houses have a very low yearly heat demand. Thus, a heat pump could be able to cover a considerable amount of the yearly heat demand in such buildings.

Some system designs are able to provide cooling of the supply air during periods with high outdoor temperatures. However, NS 3700 (2013) states that residential passive houses should achieve thermal comfort without mechanical cooling. There are usually no need for space heating or pre-heating of the supply air during the summer season, though there is still a demand for DHW. Some system solutions can heat DHW and at the same time provide cooling of the supply air.

During the same period, a heat exchanger would not recover any useful energy.

Heat pumps are dependent on electrical energy for the compressor in order to recover heat. Thus, heat pumps use energy in order to recover heat that could be recovered with no or only limited energy consumption by a ventilation heat exchanger. That is one of the major drawbacks of using a heat pump for ventilation heat recovery.

An exhaust air heat pump increases the noise level from the technical room, seeing that the compressor normally is placed within the

(39)

2.2. Heat recovery using exhaust air heat pumps 13

building envelope, as opposed to heat pumps using outdoor air as only heat source. Thus, a higher requirement for sound insulation of the AHU cabinet or the technical room is required.

2.2.1 Thermodynamic basis for CO2 heat pumps

Heat pumps are able to move energy from an energy source with low temperature to an energy sink with higher temperature, using heat exchangers, a compressor, an expansion device and a suitable refrigerant.

The most basic components are illustrated in figure 2.2.

Figure 2.2: Illustration of the most basic components in a heat pump Heat pumps operate in thermodynamic cycles. These cycles are usually drawn in log P-h diagrams or T-s diagrams. An additional diagram, the T-h diagram, is much used for CO2 heat pumps. An illustration of a T-h diagram for a transcritical CO2 heat pump cycle is shown in figure 2.3.

Dry saturated gas from the evaporator is compressed in the compressor. An ideal reversible compressor would result in condition 2is. Condition 2ad is a state representing adiabatic compression, where the isentropic efficiency is considered. The last state, 2r, is the real state

(40)

Figure 2.3: T-h diagram showing a general transcritical heat pump cycle. 2isequals isentropic compression, 2ad equals adiabatic compression, while 2r is the real compression process, including heat loss from the compressor. Heat is given off from CO2 (red line) to water (blue line). GC is short for ”gas cooler”.

of the refrigerant after compression. This process point has lower enthalpy than 2ad due to heat loss from the compressor to the surround- ings. However, the enthalpy can be both higher and lower than 2is, depending on the heat loss rate. The equations and correlations used to calculate thermodynamic states are presented in chapter 5.3.2 (p. 67).

Heat is given off from CO2to water in the gas cooler, after the compression process. As distinct from subcritical heat pumps, heat is given off at a large temperature glide. This results in a good temperature adaptation between the CO2 being cooled and the water being heated.

The difference between heat rejection in a condenser (subcritical) and a gas cooler (transcritical) is illustrated in figure 2.4. Heat rejection at a large temperature glide results in a higher COP compared to heat rejection at a constant temperature. The heat transfer properties of CO2

is also excellent (Stene, 2004). Thus, a transcritical CO2 heat pump is a good choice for heating of DHW.

CO2 differs strongly from refrigerants for subcritical heat pump

(41)

Figure 2.4: Comparison of average temperature during heat rejection for subcritical and transcritical heat pump. Ta is the temperature approach.

cycles. Subcritical cycles often use ammonia, HFCs or hydrocarbons (usually R290). The major difference between CO2 and subcritical refrigerants is the critical point. The critical point of CO2 is at 31.1 °C and 73.8 bar, making it suited for transcritical operation.

Transcritical pressure lines (isobars) for CO2are shown in figure 2.5.

Components for subcritical heat pump cycles often have a pressure rating of 25 bar to 40 bar. Some industry standard heat pumps operate at 60 bar. CO2heat pump components have a pressure rating of 120 bar to 150 bar on the high pressure side Stene (2014).

Increased gas cooler pressure results in a more linear isobar, which is advantageous for heat transfer in the gas cooler. The CO2gas behave more and more like an ideal gas at high pressures. Higher gas cooler pressures also increase the required compression work and the CO2

discharge gas temperature. Thus, finding the ideal gas cooler pressure is not easily done.

(42)

Figure 2.5: Transcritical pressure lines for CO2 in a T-h diagram.

The data was generated using NIST REFPROP.

2.2.2 System configurations

There are several different possible system configurations for heat pumps recovering energy from the exhaust air. The next sections assess the most common system configurations for exhaust air heat pumps. All configurations will experience frost formation on the evaporator when used in cold climates. Defrosting is required regularly, which reduces the SPF of the heat pump .

Air–air configuration

The air–air system configuration has an evaporator placed in the exhaust air duct and a condenser placed in the supply air duct. Heat is transferred directly from the exhaust air to the supply air. Heat cannot be stored, and all required heat must be produced instantaneously. The configuration can provide cooling of he supply air, if equipped with a four-way valve. Energy removed from the supply air is not utilized, but rejected to the exhaust air leaving the building.

(43)

The heat pump cannot be used for heating of DHW. During periods with high outdoor temperatures, the heat pump is thus not used (except for cooling). Efficient rotating wheels have efficiencies between 80% and 85%, which most of the year results in a sufficiently high supply air temperature in normal or mild Norwegian climates. An air–air exhaust air heat pump uses electrical energy in order to recover the same energy, and is thus much less efficient.

This heat pump configuration may be used for space heating. Warm air, heated by the heat pump, is supplied to the zones. Space heating through ventilation air has low investment costs compared to hydronic heating, but is not much used in Norway due to reduced ventilation efficiency and reduced thermal comfort. However, a recent study indi- cates that ventilation heating in Norwegian passive houses in could be feasible, due to the low heating demands (Holte, 2013).

Air–water configuration

Air–water exhaust air heat pumps reject the energy in the condenser or gas cooler to water. The condenser or gas cooler may transfer heat directly to a thermal storage, or to water circulating between the condenser/gas cooler and the thermal storage. Heat may also be given off directly to water in a hydronic heating circuit. The heated water may be used for DHW, space heating or ventilation heating, and is thus much more flexible than the air–air configuration.

Products with this system configuration are usually not able to provide cooling of the supply air, seeing that they do not have a heat exchanger (CO2–air) placed in the supply air duct. However, a solution proposed by Renedo et al. (2007) enables cooling of the supply air and heating of DHW at the same time. Dampers are used to switch the flow direction through the evaporator. Thus, the supply air is guided to the heat pump evaporator, while the exhaust air bypasses the evaporator.

Air–air/water configuration

This heat pump configuration is able to give off heat to both water and to the ventilation supply air. This system configuration is used by the Danish AHU and heat pump manufacturer Nilan (see chapter 3.3.3,

(44)

p. 35), among others. The advantage of air–air/water configurations is the possibility to cool the supply air and heat DHW simultaneously.

This is made possible by placing a heat exchanger (CO2–air) in the supply air duct, together with a CO2–water heat exchanger in the storage tank. A simplified sketch of this system configuration is shown in figure 2.6.

Figure 2.6: System configuration sketch for air–air/water systems in heating mode. The dotted lines illustrate the refrigerant flow in cooling mode. The illustration does not take optimum flow directions in the heat exchangers into consideration.

When running in cooling mode, a four-way valve changes the refrigerant flow through the heat exchangers in the exhaust and supply air duct. Heat is given off to the DHW tank. Any surplus energy is given off in the exhaust air duct.

This system configuration is often used for space heating through ventilation air. The basic model from Nilan (chapter 3.3.3) is only able to provide space heating through warm supply air, and not through hydronic heating in the zones.

(45)

2.3. Hybrid heat recovery 19

2.3 Hybrid heat recovery

Hybrid heat recovery solutions use a ventilation heat exchanger (chapter 2.1), often plate or heat wheel, and one of the heat pump configurations mentioned above (chapter 2.2.2). The exhaust air passes through a ventilation heat exchanger and gives off heat to the supply air. Thereafter, exhaust air passes through the heat pump evaporator before leaving the building. A ventilation heat exchanger have an efficiency of less than 100%, resulting in an air temperature after the ventilation heat exchanger above the outdoor air temperature. The difference in temperature between the discharge air after the heat exchanger and the outdoor air is illustrated in figure 2.7 for different heat exchanger efficiencies and outdoor temperatures.

Figure 2.7: Temperature difference between discharge air and outdoor air for different heat exchanger efficiencies. The indoor temperature is set to 20 °C.

The difference in temperature between the air after the ventilation heat exchanger and outdoor air is taken advantage of by the exhaust air heat pump. The increased temperature of the heat source is ben- eficial for the heat pump performance, seeing that more energy can be removed from the air before reaching the minimum temperature.

(46)

Alternatively, the heat pump will be able to operate at an increased evaporation temperature.

The products from Genvex, Nilan and Exvent (chapter 3.3, p. 31) are examples of products for hybrid heat recovery. The products from Genvex and Nilan are based on the air–air/water system configuration, using plate heat exchangers for ventilation heat recovery. The product from Exvent is based on the air–air system configuration, using a heat wheel for ventilation heat recovery. According to Fabrizio et al.

(2014), the main advantage for hybrid heat recovery systems is the packed configuration and its easy assembly and installation in residential buildings.

2.4 Heat pumps for ventilation heat recovery in NS 3031 (2014)

The revised edition of NS 3031 from 2014 contains a supplement for calculation of the performance for exhaust air heat pumps. Recovered energy from the exhaust air can be used for preheating of the supply air, but also DHW and space heating, and still be regarded asrecovered energy. Figure 2.8 illustrates the energy flows in an exhaust air heat pump.

The amount of recovered energy equals energy removed from the exhaust air down to outdoor temperature. Removal of energy from the exhaust air beneath the outdoor temperature is not regarded as recovered energy. The useful energy from the exhaust air heat pump consists of three parts:

• Recovered energy

• Compressor energy used to recover the above energy

• Energy removed from exhaust air below the outdoor temperature, plus the compressor energy used to obtain that energy

The first part, the recovered energy, is calculated according to equation 2.1. ηequals the combined efficiency for heat distribution, heating control and heat emission. Equation 2.2 calculates the amount of energy used by the compressor for heat recovery. Equation 2.3 calculates

(47)

2.4. Heat pumps for ventilation heat recovery in NS 3031 (2014) 21

Figure 2.8: Energy flows in an exhaust air heat pump, according to NS 3031 (2014).

the third share of energy from the exhaust air heat pump. The annual amount of recovered energy is thus calculated according to equation 2.4, where a time step of one hour or less should be used. The recovered energy from the exhaust air is subtracted from the annual energy demand for heating in the building. The energy used by the compressor is added to the net energy heating demand.

Q˙₁ηCOP−1

COP−1 (2.1)

Q˙₁

COP−1 (2.2)

ηQ˙2COP

COP−1 (2.3)

Q_rvd,hp=

8760

X

i=1

Q˙1,i

ηiCOPi−1

COPi−1 (2.4)

(48)

(49)

Chapter 3

Literature

Relevant literature on heat exchangers and heat pumps for ventilation heat recovery were presented in the project report (Solberg, 2014). This chapter will provide some additional literature on heat pumps for ventilation heat recovery, including hybrid heat recovery. The following scientific databases and research publication websites were used in the search for relevant literature:

• AIVC Publications

• DAIM

• DiVA portal

• Fraunhofer Publica

• Google Scholar

• Purdue e-Pubs

• REHVA Publications & Resources

• ResearchGate

• ScienceDirect

• Scopus

• Web og Science

23

(50)

24 Chapter 3. Literature

Different combinations of the following keywords were used in the search for relevant publications. In addition, the bibliography of relevant publications was looked over in search for other useful publications.

Some employees at SINTEF and NTNU were also asked for relevant publications.

• Carbon dioxide

• CO2

• CVHD

• EAHP

• Energy recovery

• Exhaust air

• Heat pump

• Heat recovery

• Ventilation air

Only a limited number of relevant publications were found. Most of the reports have investigated the performance of an exhaust air heat pump or a hybrid recovery unit at a constant heat demand and constant temperatures, for example at winter or summer conditions. No reports were found where the annual energy performance of a residential building equipped with a heat pump for heat recovery was assessed.

In the following sections, publications regarding exhaust air heat pump solutions and hybrid solutions are looked into and discussed.

3.1 Exhaust air heat pump heat recovery

Ott (2006) set up and tested an exhaust air CO2 heat pump for combined space heating and hot water heating. The heat pump used active control of both gas cooler pressure, super heating and water mass flows in order to maximize the COP. The gas cooler pressure was altered according to operating mode: tap water heating mode or space heating mode. The prototype heat pump had a heating capacity of 2 kW, removing energy from 150 m³/h exhaust air initially at 20 °C. Water

(51)

3.1. Exhaust air heat pump heat recovery 25

was heated from 25 °C to 35 °C in space heating mode. Corresponding numbers for heating of DHW were 17 °C and 70 °C. The evaporation temperature was−1 °C, while the optimum super heating was chosen in a tolerated range from 3 K to 9 K.

Results from the prototype CO2 heat pump experiments showed that the gas cooler pressure resulting in the highest COP (3.22) for space heating was 82 bar. The optimum gas cooler pressure for heating of domestic hot water was 100 bar, resulting in a COP of 3.19. Ott also compared these results to a propane (R290) heat pump. The propane heat pump achieved a COP of 4.2 and 2.7 for space heating and heating of DHW respectively.

The space heating and DHW heating COP values are approximately the same. The inlet water temperature in the case of space heating is lower than the corresponding value for heating of DHW, but the required compression work for heating of DHW is higher due to higher gas cooler pressure. These two factors seems to level each other out.

The experiments performed by Ott show that a CO2 heat pump is a rational choice for heating of DHW, but not necessarily for space heating. Three blocks of flats located in Oslo (Tveita housing coopera- tive) were recently refurbished, including the DHW and space heating plant (Borge, 2014). A large CO2 heat pump was installed for heating of DHW, and an additional heat pump using R134a was installed in order to cover the space heating demands. Both heat pumps were using exhaust air as heat source. The dedicated DHW heat pump achieved a SPF of 4.5. The performance of the R134a heat pump was not investigated.

These two reports show that a CO2heat pump is an excellent choice for heating of DHW, though not necessarily for space heating. However, Stene (2004) looked into a CO2 heat pump forcombined space heating and heating of DHW, and compared it to a heat pump using a HFC as refrigerant. The results showed that the CO2 heat pump performed better than the HFC heat pump in cases where the yearly delivered energy for DHW production was more than 25% to 30% of the total yearly delivered heat from the heat pump. The main reason for the high CO2 heat pump performance was the use of a tripartite gas cooler. A tripartite gas cooler consists off three parts. The last part of the gas

(52)

cooler gives off heat to preheat the DHW, the middle part gives off heat to space heating, and the third part reheats the DHW to the setpoint temperature. The tripartitioning results in an excellent temperature adaptation for water being heated and CO2 being cooled down. The COP for combined mode (DHW and space heating) was thus higher than the COP for heating of DHW only. For a single-unit dwelling, it would normally not be economically feasible to install one CO2 heat pump for heating of DHW and another HFC based heat pump for space heating. Therefore, a solution with a tripartite gas cooler should be used – both with regard to life-cycle costs and environmental aspects.

Stavset et al. (2014) performed simulations on a CO2 exhaust air heat pump for DHW, space heating and heating of ventilation air. The simulations were carried out for 5 consecutive winter days with an outdoor temperature of −10 °C. The heat pump used in the simulations consisted of a standard evaporator, suction gas heat exchanger, tripartite gas cooler, compressor and expansion valve. The gas cooler pressure was set to 85 bar constantly. The ventilation air flow rate was set to 2.1 m³/hm² when occupied, and 1.2 m³/hm²when not occupied.

Five different scenarios with different occupant behaviour models and heat pump controllers were assessed, where the DHW, SH and VH energy demands were set to constant values, depending on the occupant behaviour (home, not home or sleeping). The resulting COP for the different scenarios ranged from 4.1 to 4.5 for combined DHW, space and ventilation heating on a winter day. The heat pump was able to cover the entire heating demand (DHW, SH and VH) for the building.

The simulations were only performed for a constant outdoor climate, resulting in a constant heat demand depending on whether the occupants are home and awake, not at home, or sleeping. The model does therefore not assess the performance during different seasons. The average daily ventilation air flow rates used in the model vary from 1.8 m³/hm² to 2.1 m³/hm². This is 50% to 75% more than the minimum air flow rate according to the Norwegian building code (Technical Regulations, 2010), which may increase the COP of the heat pump, but also increase the total demand of delivered energy to the building.

Based on the numbers stated in the report, the average outlet air temperature from the evaporator is equal to about −8 degC or less

(53)

3.2. Hybrid heat recovery 27

(for scenario #1). However, the report concludes that there was no need for active defrosting of the heat pump evaporator. This is caused by periods with a lower heat demand, resulting in outlet temperatures from the evaporator above the freezing point (Alonso, 2015).

The results from the report show that an exhaust air CO2 heat pump can obtain high COP values when operating in winter mode. In addition, the report shows that the system losses due to defrosting of the evaporator may be low or non-existent.

3.2 Hybrid heat recovery

Kluge (2008) investigated the performance of ”Vitotres 343”, a CVHD (compact ventilation and heating device) from Viessmann. The device was the first of its sort from Viessmann for residential passive houses (Viessmann, 2014), and could use both exhaust air and outdoor air as heat source for the heat pump. The refrigerant used in the heat pump was R134a. Measurements on the device were carried out with an outdoor temperature of 3.5 °C to 5°C and a room temperature of 26 °C. The supply air temperature from the CVHD was 22.5 °C. The CVHD operated with an exhaust and supply air flow rate of about 137 m³/h. The fan supplying the evaporator with outdoor air was not activated during the measurements. Therefore the measurements were performed with exhaust air as only heat source. The measurements showed an heat exchanger efficiency of 83.3%. The COP of the heat pump was measured to 2.06. At these conditions, the heat pump provided a heating power of 1.28 kW.

Kluge (2008) also looked into different ways of improving the performance of the heat pump. Use of a suction gas heat exchanger (SGHX) was estimated to increase heat pump COP by 2.8% at summer conditions and 3.2% at winter conditions for the R134a heat pump. The performance is increased, but the investment cost of the CVHD is also increased. Stene (2004) performed calculations on a CO2 heat pump showing that an increasing superheat of the CO2 before compression had a very limited impact on the COP. Thus, a suction gas heat exchanger should only be used in cases where it is necessary to evaporate liquid CO2 in the suction line, in order to avoid compressor failure.

(54)

Kluge (2008) also estimated that substituting R134a with CO2

would increase the COP by 18% at summer conditions and 9.8% at winter conditions, which substantiates the advantage of using CO2 as refrigerant in such applications.

Calay and Wang (2013) proposed a hybrid heat recovery system consisting of an energy wheel and a reversible air–air heat pump. Isen- tropic compression and a constant temperature difference of 15 °C in heat exchangers were assumed. Between 64% and 73% of the heat energy could be saved using the energy wheel and the heat pump, compared to a system with an electric heater. Though not stated clearly in the report, it is assumed that it is the space heating and ventilation heating demand that are reduced by the given percentage.

The solution proposed by Calay is not able to reduce the amount of delivered energy for heating of DHW. The total energy saving is thus limited in passive houses with a low energy consumption for space heating and ventilation heating, compared to energy for heating of DHW.

The calculations performed in the report are of limited accuracy and value due to isentropic compression, very simplified heat exchanger modelling and calculations only for selected operating conditions. The choice of refrigerant is not stated.

Fraunhofer ISE proposed a CVHD solution for high performance buildings, for space heating, heating of DHW and heating of the ventilation air (Bühring, 2005). The solution provides space heating using hot ventilation supply air. The CVHD uses a heat exchanger for passive ventilation heat recovery, and is thus a hybrid solution. The Fraunhofer CVHD also had an earth–air heat exchanger to preheat the ventilation air, and a solar collector to heat the storage tank whenever possible. Such solutions have a market share of 30% to 50% in the German marked for residential passive houses and low energy houses, according to the report.

Three different commercialized products were tested according to the NS-EN 255-3 (2008) test procedure (now replaced by NS-EN 16147 (2008)). Product A had a solar collector and an earth–air heat exchanger, in addition to a passive air–air heat exchanger and an exhaust air heat pump. One condenser was located in the storage tank and another condenser was placed in the supply air duct. Product B was the

Heat Recovery in Combination with Different Heat Pump Solutions for Energy Supply